Pancreatic Morphology and Function in Adult Patients with Cystic Fibrosis
6
E. H. J . I N HAREN, W. P. M. HOPMAN, G. ROSENBUSCH, J. B. M. J . JANSEN & C. L. A. CAN HERWAARDEN Dept. of Gastroenterology and Hepatology, Dept. of Pulmonary Diseases, and Dept. of Radiology. University Hospital Nijmegen, Nijmegen, The Netherlands
Scand J Gastroenterol Downloaded from informahealthcare.com by TIB/UB Hannover on 01/10/15 For personal use only.
Van Haren EHJ, Hopman WPM, Rosenbusch G, Jansen JBMJ, Van Herwaarden CLA. Pancreatic morphology and function in adult patients with cystic fibrosis. Scand J Gastroenterol 1992;27:695-698. To determine the relation between pancreatic morphology and pancreatic exocrine and endocrine K n c t i o n , we have studied 8 adult cystic fibrosis patients and 14 normal control subjects by ultrasonography and pancreatic function testing. In the patients with cystic fibrosis the maximum anteroposterior diameter of the pancreatic head was significantly increased over that in control subjects ( p < 0.01), whereas the maximum diameter of the body was significantlydecreased ( p = 0.05). Increased echogenicity of the pancreatic body was observed in most patients. In the cystic fibrosis patients postprandial insulin secretion was reduced in the 1st h ( p < 0.005 versus control), whereas pancreatic polypeptide secretion was virtually abolished for at least 3 h ( p < 0.01 versus control). All cystic fibrosis patients had evidence of exocrine pancreatic dysfunction as reflected by a diminished urinary paraaminohenzoic acid excretion. Intraduodenal enzyme and bicarbonate output in response to secretincholecystokinin was reduced in all of three patients studied. It is concluded that loss of endocrine and exocrine pancreatic function in adult cystic fibrosis patients is accompanied by a small and echo-dense pancreatic body relative to a large pancreatic head
1
Key words: Cystic fibrosis; insulin; pancreas; pancreatic function; pancreatic polypeptide; para-amino-
benzoic acid; steatorrhea; ultrasonography
Wim P . M . Hopman, M . D . , Ph. D . , Dept. of Gastroenterology and Hepatology, University Hospital Nijmegen, P . 0. Box 9101, 6500 H B Nijmegen, The Netherlands
Improvement in diagnosis and therapy has lengthened life expectancy of children with cystic fibrosis (CF), and most of these patients now survive at least into adolescence (1). The pulmonary disease dominates the clinical picture, but pancreatic involvement is responsible for exocrine and endocrine dysfunction in the adult population as well. Most adult C F patients have severe pancreatic insufficiency. However, approximately 15% have normal fecal fat excretion (2). Diabetes mellitus occurs in about 13% of adults with C F (2). In the present study in adult CF patients pancreatic exocrine and endocrine function and morphologic characteristics of thc pancreas as determined by ultrasound were compared with those in a group of normal control subjects. Detection of functional and morphologic pancreatic disturbances provides further understanding of the pathologic process in this hereditary disease. Furthermore, the finding of pancreatic abnormalities can support the diagnosis of C F in patients with lung disease, suspected of having C F but with noninformative or borderline sweat test results (3).
SUBJECTS A N D METHODS Eight patients (six male, two female) with C F were enrolled in the study. The mean age of the patients was 28 years (range, 22-41 years). All patients had proven CF, confirmed
by a sweat chloride level of >70 mmol/l. Six patients had an increased fecal fat excretion (range, 27-67 g/24 h), whereas two patients did not have steatorrhea on repeated tests. None of the CF patients had diabetes mellitus. Fourteen age- and sex-matched healthy volunteers acted as control subjects. Informed consent was obtained from all subjects studied, and the protocol was approved by the ethics committee of Nijmegen University Hospital. After an overnight fast ultrasound examination was performed using a real-time scanner with a 3.5-MHz transducer. Images of the pancreas were obtained in accordance with the method described by Niederau et al. (4). T h e maximum anteroposterior diameter of the head and body of the pancreas was measured on a transverse oblique scan, using the upper abdominal blood vessels as landmarks (4). The patients were positioned for optimal visualization of the pancreas, including supine and decubitus position. Seven patients and eight normal subjects ingested, during a 15-min period, a standard breakfast consisting of 30g fat, 30g protein, and 70 g carbohydrate. Two millimoles of the synthetic peptide N-benzoyl-L-tyrosyl-p-aminobenzoicacid (BT-PABA) was added to the meal. Blood was drawn for measurement of glucose, insulin, and pancreactic polypeptide before and at 15-min intervals until 180min after start of the meal. After the meal, patients were not allowed
696
E. IJ. .I. rmn Huren et ul.
Pancreatic head - Pancreatic body (mm 1 ( m m1 LO LO
L
30
30
3
20
20
2
10
10
1
normal CF subjects patients
Scand J Gastroenterol Downloaded from informahealthcare.com by TIB/UB Hannover on 01/10/15 For personal use only.
-Ratio head and body-
normal CF subjects patients
normal CF subjects patients
Fig. I . Maximum anteroposterior diameter of the pancreatic head (left panel), the pancreatic hody (middle panel), and the ratio of these dimensions (right panel) in 8 cystic fibrosis (CF)patients (hatched bars) and in 14 normal subjects (open bars). Error bars denote 1 SEM.
to eat but were asked to drink at least 1.5 Iwater to guarantee sufficient urine production, which was collected for 6 h after the meal. The recovery of para-aminobenzoic acid (PABA) was determined in the 6-h urine. Plasma insulin and pancreatic polypeptide (PP) were measured by radioimmunoassay (5,6) and PABA in urine by photometry (7). A secretin--cholecystokinin stimulation test was performed in three patients, two with overt steatorrhea and one without steatorrhea. The duodenal output of amylase, trypsin, and bicarbonate was measured and compared with normal values from our laboratory (8). Unless otherwise stated, results were expressed as the mean ? 1 SD. Basal values were calculated by taking the mean of two fasting samples. Integrated plasma insulin secretion and integrated plasma PP secretion were determined by calculating the area under the plasma concentration time curve, after subtraction of the basal value. Statistical analysis was performed with Student’s t test and the Mann-Whitney U-test. Spearman’s rank order correlation coefficient was used to determine the correlation between pancreatic size measurements and the other parameters studied. All tests were two-tailed.
Integrated plasma PP(nmol/L. 60rnin)
m pc 0.0005
Integrated plasma lnsul i n ( mU/ L ,60rnin 1600 I NS
$‘
8 00 T
*
*
P(001
LL I
NS--
O Oo5’
LOO
*
-
1200
RESULTS The maximum anteroposterior diameter of the pancreatic head in C F (31.0 9.1 mm) was significantly larger than that of control subjects (22.5 4.9 mm; p < 0.01) (Fig. l), whereas pancreatic body diameter was significantly smaller (11.4 Ifr 1.7 mm in C F versus 13.0 If: 1.9 mm in control subjects; p = 0.05) (Fig. 1). The ratio of the pancreatic head to body diameter was significantly greater in the C F patients (2.8 Ifr 1.1) than in control subjects (1.8 0.4; p < 0.005) (Fig. 1). Ultrasonography of the pancreas showed an increased echogenicity in five of the seven patients. In one CF patient without a history of pancreatitis a homogeneous anechoic cavity with a diameter of 2 cm, compatible with a small cyst, was found in the head of the pancreas. The basal plasma PP level in CF (12 k 3 pmol/l) was
-
CF
patients 60 min
f
1
norrml CF subjects patients 60-120min
normal CF su bjeCts patients 120 - 180min
0Fig. 2. Hourly integrated plasma pancreatic polypeptide (PP) secretion (upper panel) and hourly integrated plasma insulin secretion (lower panel) in 7 cystic fibrosis (CF) patients (hatched bars) and in 8 normal subjects (open bars). Error bars denote 1 SEM.
significantly ( p < 0.05) lower than that in the control subjects (24 2 14 pmol/l). In the control subjects meal ingestion induced an immediate and statistically significant ( p < 0.001) increase of the plasma PP level to a peak value of
Pancreas in Cystic Fibrosis
Table I. Secretin-cholecystokinin stimulation test Patient
Faecal fat excretion k / 2 4 h)
Scand J Gastroenterol Downloaded from informahealthcare.com by TIB/UB Hannover on 01/10/15 For personal use only.
Maximum amylase concentration (% of normal) Maximum trypsin concentration (% of normal) Bicarhonate excretion (% or normal)
A
B
C
67.1
27.4
2.7
1.2
0.3
42
1.9
0.3
84
12.5
7.0
4.0
106 -+ 34 pmol/l at 75 min. In the CF patients, however, meal ingestion did not induce any significant change in plasma PP. Integrated plasma PP secretion in the CF patients was significantly lower than that in the control subjects until 3 h after the meal ( p < 0.01) (Fig. 2, upper panel). The basal plasma insulin concentration did not differ significantly between the CF patients (7.6 3.7 mu/]) and the normal subjects (11.3 ? 2.0 mU/l; p = 0.08). However, integrated plasma insulin secretion in response to meal ingestion was significantly lower during the 1st h ( p < 0.005) (Fig. 2 , lower panel). This difference subsided in the 2nd and 3rd h after the meal (Fig. 2, lower panel). Basal (4.2 C 0.8 mmol/l) and maximum plasma glucose concentrations (8.7 2.1 mol/l) in CF patients were similar to those in normal subjects (3.9 t 0.6mmol/l and 7.6 -t 3.1 mmol/l, respectively; NS). Urinary PABA excretion was reduced in all CF patients (19 +- 13%; range, 8-4376) when compared with normal subjects (59 17%; range, 5 1 4 8 % ; p < 0.005). The pancreatic body diameter was significantly correlated with urinary PABA excretion ( p < 0.01), basal plasma PP concentration ( p < 0.05), and maximum postprandial plasma PP level ( p < 0.05). The results of the secretincholecystokinin stimulation test in three patients with CF are summarized in Table I .
*
*
*
DISCUSSION Ultrasonographic examination of the pancreas in CF patients showed a larger pancreatic head and a smaller pancreatic body than in normal subjects. Furthermore, the ratio of pancreatic head to body diameter in CF patients was much greater than that in normal subjects. The reason for these morphologic characteristics of the pancreas in CF patients is not known. In chronic pancreatitis enlargement of the pancreas occurs initially, and later in the course atrophy is seen (9). Theoretically, it may be possible that the pathologic process in CF has a more severe damaging effect on the body initially, sparing the pancreatic head. During the development ofdamage to the pancreatic body, the pancreatic head may increase to compensate for loss of functioning pancreatic tissue. The finding of an increased echogenicity
691
of the pancreatic body in most CF patients, which has been attributed to fibrosis and fatty replacement (10, l l ) , together with the finding that selenomethionine (’%e) scanning of the pancreas in CF patients showed uptake only in the pancreatic head (12) supports this hypothesis. In one patient an anechoic cavity 2 cm in diameter, which was probably a cyst, was seen on sonography. Pancreatic cystosis is, like fibrosis and steatosis, a characteristic pathologic feature, and the name ‘cystic fibrosis’ is derived from these hallmarks (13). Pancreatic cysts in CF are true cysts, lined by a single layer of epithelium and containing inspissated eosinophilic material. The defective bicarbonate, chloride, and fluid transport is thought to lead to intrapancreatic protein hyperconcentration stasis and secondary ductal ectasia (14). The pancreatic cysts can cause compression and pain in CF patients and may occasionally require surgical therapy. Both in patients with and without steatorrhea exocrine pancreatic function, as determined by the NBT-PABA test and the secretin-cholecystokinin test, was disturbed. The secretin-cholecystokinin stimulation test in the CF patient without steatorrhea showed a moderately reduced pancreatic enzyme output but a strongly reduced bicarbonate output. Steatorrhea does not seem to occur until the secretion of pancreatic lipase is less than approximately 10% of normal, but many other factors appear to play a role, such as lingual lipase activity and co-lipase deficiency (15,16). The finding of low duodenal bicarbonate output after secretin stimulation indicates a discrepancy between acinar (enzymes) and ductular (fluid, bicarbonate, electrolytes) pancreatic function, not often seen in other forms of pancreatic insufficiency (2). The CF patients, regardless of their ability to produce adequate pancreatic enzyme output, have an abnormality of pancreatic fluid and bicarbonate secretion related to the genetically defective chloride channel function (17, 18). To determine this discrepancy between acinar and ductular function, both secretin and cholecystokinin stimulation is necessary in pancreatic function testing. All CF patients had evidence of an impaired endocrine pancreatic function as reflected by a blunted insulin secretion and virtual absence of PP secretion in response to a standard meal. This finding is in agreement with a previous study in adult CF patients (19). PP-producing cells occur almost exclusively in the pancreas (20). Postprandial increase in plasma PP has been suggested to be a measure of functional pancreatic mass in chronic pancreatitis, but in these patients the PP secretion remained normal until the pancreas was so damaged that the disease usually had revealed itself by overt steatorrhea (20). PP cells are scattered in the exocrine parenchyma and in the islets, where they are usually located at the periphery (20,21). In the present study basal and maximal plasma PP levels correlated well with the diamter of the pancreatic body, suggesting that pancreatic atrophy is accompanied by a decrease in either the number or function of PP cells. Disruption of pancreatic architecture by
Scand J Gastroenterol Downloaded from informahealthcare.com by TIB/UB Hannover on 01/10/15 For personal use only.
698
E. H . J . van Haren et al.
fibrosis and fatty infiltration is probably responsible for the observed findings (13). A blunted but still significant insulin response to meal ingestion was found irrespective of exocrine pancreatic function. The results therefore indicate that the insulin-producing cells located in the islets are much less susceptible to the damaging process of the pancreas in CF than the PP-producing cells and the exocrine tissue. Nevertheless, the number of islets per cubic millimeter of pancreatic tissue was decreased in CF patients with and without diabetes (22). In addition, it should be noted that islet cell antibodies may also play a role in the development of diabetes mellitus in CF (23). The sweat test is the cornerstone in the diagnosis of CF, but the test can yield borderline results. Modification of the sweat test by means of fluorocortisone can aid the diagnosis further (24), but normal sweat electrolyte levels may be found in patients with typical CF (25). In such cases ancillary criteria for the diagnosis of CF have to be relied on. The demonstration of characteristic features at ultrasound and at exocrine pancreatic function testing can support the diagnosis of CF. It is concluded that in adult CF patients ultrasonography demonstrates abnormalities in pancreatic morphology which are characterized by a large pancreatic head relative to a small and echo-dense body. Exocrine and endocrine pancreatic function are disturbed in all patients, irrespective of the presence of steatorrhea. These functional and morphologic abnormalities can support the diagnosis of CF in case of borderline or negative sweat test results.
REFERENCES 1. David TJ. Cystic fibrosis. Arch Dis Child 1990;65:152-7. 2. Park RW, Grand RJ. Gastrointestinal manifestations of cystic fibrosis: a review. Gastroenterology 1981;81:114M1. 3. Hall SK, Stableforth DE, Green A. Sweat sodium and chloride eoneentrations-essential criteria for the diagnosis of cystic fibrosis in adults. Ann Clin Biochem 1990;27:318-20. 4. Niederau C, Sonnenberg A, Muller JE, Eckenbrecht JF, Scholten T, Fritsch WP. Sonographic measurements of the normal liver, spleen, pancreas and portal vein. Radiology 1Y83; 149:53740. 5. Yalow RS, Bcrson SA. Immunoassay of endogenous plasma insulin in man. J Clin Invest 1960;39:1157-75. 6. Lamers CBHW, Diemel CM, Jansen JBMJ. Comparative study of plasma pancreatic polypeptide responses to food, secretin
Received 27 December 1991 Accepted 3 March 1992
and bombesin in normal subject and in patients with chronic pancreatitis. Dig Dis Sci 1984;29: 102-8. 7. Bratton AC, Marshall EK. A new coupling component for sulfonimide determination. J Biol Chem 1939;128:537-50, 8. Lankisch PG. Exocrine pancreatic function tests. Gut 1982; 23: 777-98. 9. MacCarthy RL, Word EM, Charboncau JW, Sheedy PF, May GR. Imaging of the pancreas. In: Go VLW, et al., editors. The exocrine pancreas: biology, pathobiology and diseases. New York: Raven Prcss, 1986:841-72. 10. Graham N , Manhire AR, Stead RJ, Lees WR, Hodson ME, Batten JC. Cystic fibrosis: ultrasonographic findings in the pancreas and hepatobiliary system correlated with clinical data and pathology. Clin Radio1 1985;36: 199-203. 11. Phillips HE, Cox CL, Reid MH, McGahan JP. Pancreatic sonography in cystic fibrosis. AJR 1981:137:6Y-72. 12. Schwachrnan 11, Lebenthal E, Khaw KT. Recurrent acute pancreatitis in patients with cystic fibrosis with normal pancreatic enzymes. Pediatrics 1975;55:86-95. 13. Lebenthal E, Lerner A, Heitlinger L. The pancreas in cystic fibrosis. In: Go VLW, et al., editors. The exocrine pancreas: biology, pathobiology and diseases. New York: Raven Press, 1986:783-X 17. 14. Hernanz M , Teele RL, Perez A, et al. Pancreatic cystosis i n cystic fibrosis. Radiology 1986;158:629-31. IS. Fredrikson B, Blackberg L. Lingual lipase: an important lipase in the digestion of dietary lipids in cystic fibrosis. Pediatr Rcs 1980; 14: 1387-90. 16. Gaskin KJ, Durie PR, Lee L. Colipase and lipase secretion in childhood-onset pancreatic insufficiency. Gastroenterology 1984;86:1-7. 17. Gaskin KJ, Durie PR, Corey M, Wei P, Forstner G. Evidence for a primary defcct of pancreatic HCO,-secretion in cystic fibrosis. Pcdiatr Res 1982;16:554-7. 18. Davidson GP, Kirubakaran CP, Ratcliffe G , Cooper DM, Robb TA. Abnormal pancreatic electrolyte secretion in cystic fibrosis. Acta Pacdiatr Scand 1986;75:145-50. 19. Allen JM, Penketh ARL, Adrian TE, et al. Adult cystic fibrosis: postprandial rcsponse of gut regulatory peptides. Gastroenterology 1983;85:1379-83. 20. Schwartz TW. Pancreatic polypeptide: a hormone under vagal control. Gastroenterology 1983;85: 1411-25. 21. Larsson L1, Sundler F, Hakanson R. Pancreatic polypeptidea postulated new hormone: identification of its cellular storage site by light and electron microscopic immunocytochcrnistry. Diabetologia 1976; 12:211-26. 22. Rodman HA, Docrshuli CF, Roland JM. The interaction of two diseases: diabetes mellitus and cystic fibrosis. Medicine 1986;65:389-97. 23. Stutchlield PR, O’Halloran SM, Smith CS, Woodrow JC, Bottazzo GF, Heaf 0. HLA-typc, islet cell antibodies and glucose intolerance in cystic fibrosis. Arch Dis Child 1988;63: 1234-9. 24. Hodson ME, Beldon J, Power R , Duncan F. Bambcr M, Batten JC. Sweat tests to diagnose cystic fibrosis in adults. Br Med J 1983i286: 1381-3. 25. Davis PB, Hubbard VS, Di Sant’Agnese PA. Low sweat electrolytes in a patient with cystic fibrosis. Am J Med 1980; 69:643-6.
6
E. H. J . I N HAREN, W. P. M. HOPMAN, G. ROSENBUSCH, J. B. M. J . JANSEN & C. L. A. CAN HERWAARDEN Dept. of Gastroenterology and Hepatology, Dept. of Pulmonary Diseases, and Dept. of Radiology. University Hospital Nijmegen, Nijmegen, The Netherlands
Scand J Gastroenterol Downloaded from informahealthcare.com by TIB/UB Hannover on 01/10/15 For personal use only.
Van Haren EHJ, Hopman WPM, Rosenbusch G, Jansen JBMJ, Van Herwaarden CLA. Pancreatic morphology and function in adult patients with cystic fibrosis. Scand J Gastroenterol 1992;27:695-698. To determine the relation between pancreatic morphology and pancreatic exocrine and endocrine K n c t i o n , we have studied 8 adult cystic fibrosis patients and 14 normal control subjects by ultrasonography and pancreatic function testing. In the patients with cystic fibrosis the maximum anteroposterior diameter of the pancreatic head was significantly increased over that in control subjects ( p < 0.01), whereas the maximum diameter of the body was significantlydecreased ( p = 0.05). Increased echogenicity of the pancreatic body was observed in most patients. In the cystic fibrosis patients postprandial insulin secretion was reduced in the 1st h ( p < 0.005 versus control), whereas pancreatic polypeptide secretion was virtually abolished for at least 3 h ( p < 0.01 versus control). All cystic fibrosis patients had evidence of exocrine pancreatic dysfunction as reflected by a diminished urinary paraaminohenzoic acid excretion. Intraduodenal enzyme and bicarbonate output in response to secretincholecystokinin was reduced in all of three patients studied. It is concluded that loss of endocrine and exocrine pancreatic function in adult cystic fibrosis patients is accompanied by a small and echo-dense pancreatic body relative to a large pancreatic head
1
Key words: Cystic fibrosis; insulin; pancreas; pancreatic function; pancreatic polypeptide; para-amino-
benzoic acid; steatorrhea; ultrasonography
Wim P . M . Hopman, M . D . , Ph. D . , Dept. of Gastroenterology and Hepatology, University Hospital Nijmegen, P . 0. Box 9101, 6500 H B Nijmegen, The Netherlands
Improvement in diagnosis and therapy has lengthened life expectancy of children with cystic fibrosis (CF), and most of these patients now survive at least into adolescence (1). The pulmonary disease dominates the clinical picture, but pancreatic involvement is responsible for exocrine and endocrine dysfunction in the adult population as well. Most adult C F patients have severe pancreatic insufficiency. However, approximately 15% have normal fecal fat excretion (2). Diabetes mellitus occurs in about 13% of adults with C F (2). In the present study in adult CF patients pancreatic exocrine and endocrine function and morphologic characteristics of thc pancreas as determined by ultrasound were compared with those in a group of normal control subjects. Detection of functional and morphologic pancreatic disturbances provides further understanding of the pathologic process in this hereditary disease. Furthermore, the finding of pancreatic abnormalities can support the diagnosis of C F in patients with lung disease, suspected of having C F but with noninformative or borderline sweat test results (3).
SUBJECTS A N D METHODS Eight patients (six male, two female) with C F were enrolled in the study. The mean age of the patients was 28 years (range, 22-41 years). All patients had proven CF, confirmed
by a sweat chloride level of >70 mmol/l. Six patients had an increased fecal fat excretion (range, 27-67 g/24 h), whereas two patients did not have steatorrhea on repeated tests. None of the CF patients had diabetes mellitus. Fourteen age- and sex-matched healthy volunteers acted as control subjects. Informed consent was obtained from all subjects studied, and the protocol was approved by the ethics committee of Nijmegen University Hospital. After an overnight fast ultrasound examination was performed using a real-time scanner with a 3.5-MHz transducer. Images of the pancreas were obtained in accordance with the method described by Niederau et al. (4). T h e maximum anteroposterior diameter of the head and body of the pancreas was measured on a transverse oblique scan, using the upper abdominal blood vessels as landmarks (4). The patients were positioned for optimal visualization of the pancreas, including supine and decubitus position. Seven patients and eight normal subjects ingested, during a 15-min period, a standard breakfast consisting of 30g fat, 30g protein, and 70 g carbohydrate. Two millimoles of the synthetic peptide N-benzoyl-L-tyrosyl-p-aminobenzoicacid (BT-PABA) was added to the meal. Blood was drawn for measurement of glucose, insulin, and pancreactic polypeptide before and at 15-min intervals until 180min after start of the meal. After the meal, patients were not allowed
696
E. IJ. .I. rmn Huren et ul.
Pancreatic head - Pancreatic body (mm 1 ( m m1 LO LO
L
30
30
3
20
20
2
10
10
1
normal CF subjects patients
Scand J Gastroenterol Downloaded from informahealthcare.com by TIB/UB Hannover on 01/10/15 For personal use only.
-Ratio head and body-
normal CF subjects patients
normal CF subjects patients
Fig. I . Maximum anteroposterior diameter of the pancreatic head (left panel), the pancreatic hody (middle panel), and the ratio of these dimensions (right panel) in 8 cystic fibrosis (CF)patients (hatched bars) and in 14 normal subjects (open bars). Error bars denote 1 SEM.
to eat but were asked to drink at least 1.5 Iwater to guarantee sufficient urine production, which was collected for 6 h after the meal. The recovery of para-aminobenzoic acid (PABA) was determined in the 6-h urine. Plasma insulin and pancreatic polypeptide (PP) were measured by radioimmunoassay (5,6) and PABA in urine by photometry (7). A secretin--cholecystokinin stimulation test was performed in three patients, two with overt steatorrhea and one without steatorrhea. The duodenal output of amylase, trypsin, and bicarbonate was measured and compared with normal values from our laboratory (8). Unless otherwise stated, results were expressed as the mean ? 1 SD. Basal values were calculated by taking the mean of two fasting samples. Integrated plasma insulin secretion and integrated plasma PP secretion were determined by calculating the area under the plasma concentration time curve, after subtraction of the basal value. Statistical analysis was performed with Student’s t test and the Mann-Whitney U-test. Spearman’s rank order correlation coefficient was used to determine the correlation between pancreatic size measurements and the other parameters studied. All tests were two-tailed.
Integrated plasma PP(nmol/L. 60rnin)
m pc 0.0005
Integrated plasma lnsul i n ( mU/ L ,60rnin 1600 I NS
$‘
8 00 T
*
*
P(001
LL I
NS--
O Oo5’
LOO
*
-
1200
RESULTS The maximum anteroposterior diameter of the pancreatic head in C F (31.0 9.1 mm) was significantly larger than that of control subjects (22.5 4.9 mm; p < 0.01) (Fig. l), whereas pancreatic body diameter was significantly smaller (11.4 Ifr 1.7 mm in C F versus 13.0 If: 1.9 mm in control subjects; p = 0.05) (Fig. 1). The ratio of the pancreatic head to body diameter was significantly greater in the C F patients (2.8 Ifr 1.1) than in control subjects (1.8 0.4; p < 0.005) (Fig. 1). Ultrasonography of the pancreas showed an increased echogenicity in five of the seven patients. In one CF patient without a history of pancreatitis a homogeneous anechoic cavity with a diameter of 2 cm, compatible with a small cyst, was found in the head of the pancreas. The basal plasma PP level in CF (12 k 3 pmol/l) was
-
CF
patients 60 min
f
1
norrml CF subjects patients 60-120min
normal CF su bjeCts patients 120 - 180min
0Fig. 2. Hourly integrated plasma pancreatic polypeptide (PP) secretion (upper panel) and hourly integrated plasma insulin secretion (lower panel) in 7 cystic fibrosis (CF) patients (hatched bars) and in 8 normal subjects (open bars). Error bars denote 1 SEM.
significantly ( p < 0.05) lower than that in the control subjects (24 2 14 pmol/l). In the control subjects meal ingestion induced an immediate and statistically significant ( p < 0.001) increase of the plasma PP level to a peak value of
Pancreas in Cystic Fibrosis
Table I. Secretin-cholecystokinin stimulation test Patient
Faecal fat excretion k / 2 4 h)
Scand J Gastroenterol Downloaded from informahealthcare.com by TIB/UB Hannover on 01/10/15 For personal use only.
Maximum amylase concentration (% of normal) Maximum trypsin concentration (% of normal) Bicarhonate excretion (% or normal)
A
B
C
67.1
27.4
2.7
1.2
0.3
42
1.9
0.3
84
12.5
7.0
4.0
106 -+ 34 pmol/l at 75 min. In the CF patients, however, meal ingestion did not induce any significant change in plasma PP. Integrated plasma PP secretion in the CF patients was significantly lower than that in the control subjects until 3 h after the meal ( p < 0.01) (Fig. 2, upper panel). The basal plasma insulin concentration did not differ significantly between the CF patients (7.6 3.7 mu/]) and the normal subjects (11.3 ? 2.0 mU/l; p = 0.08). However, integrated plasma insulin secretion in response to meal ingestion was significantly lower during the 1st h ( p < 0.005) (Fig. 2 , lower panel). This difference subsided in the 2nd and 3rd h after the meal (Fig. 2, lower panel). Basal (4.2 C 0.8 mmol/l) and maximum plasma glucose concentrations (8.7 2.1 mol/l) in CF patients were similar to those in normal subjects (3.9 t 0.6mmol/l and 7.6 -t 3.1 mmol/l, respectively; NS). Urinary PABA excretion was reduced in all CF patients (19 +- 13%; range, 8-4376) when compared with normal subjects (59 17%; range, 5 1 4 8 % ; p < 0.005). The pancreatic body diameter was significantly correlated with urinary PABA excretion ( p < 0.01), basal plasma PP concentration ( p < 0.05), and maximum postprandial plasma PP level ( p < 0.05). The results of the secretincholecystokinin stimulation test in three patients with CF are summarized in Table I .
*
*
*
DISCUSSION Ultrasonographic examination of the pancreas in CF patients showed a larger pancreatic head and a smaller pancreatic body than in normal subjects. Furthermore, the ratio of pancreatic head to body diameter in CF patients was much greater than that in normal subjects. The reason for these morphologic characteristics of the pancreas in CF patients is not known. In chronic pancreatitis enlargement of the pancreas occurs initially, and later in the course atrophy is seen (9). Theoretically, it may be possible that the pathologic process in CF has a more severe damaging effect on the body initially, sparing the pancreatic head. During the development ofdamage to the pancreatic body, the pancreatic head may increase to compensate for loss of functioning pancreatic tissue. The finding of an increased echogenicity
691
of the pancreatic body in most CF patients, which has been attributed to fibrosis and fatty replacement (10, l l ) , together with the finding that selenomethionine (’%e) scanning of the pancreas in CF patients showed uptake only in the pancreatic head (12) supports this hypothesis. In one patient an anechoic cavity 2 cm in diameter, which was probably a cyst, was seen on sonography. Pancreatic cystosis is, like fibrosis and steatosis, a characteristic pathologic feature, and the name ‘cystic fibrosis’ is derived from these hallmarks (13). Pancreatic cysts in CF are true cysts, lined by a single layer of epithelium and containing inspissated eosinophilic material. The defective bicarbonate, chloride, and fluid transport is thought to lead to intrapancreatic protein hyperconcentration stasis and secondary ductal ectasia (14). The pancreatic cysts can cause compression and pain in CF patients and may occasionally require surgical therapy. Both in patients with and without steatorrhea exocrine pancreatic function, as determined by the NBT-PABA test and the secretin-cholecystokinin test, was disturbed. The secretin-cholecystokinin stimulation test in the CF patient without steatorrhea showed a moderately reduced pancreatic enzyme output but a strongly reduced bicarbonate output. Steatorrhea does not seem to occur until the secretion of pancreatic lipase is less than approximately 10% of normal, but many other factors appear to play a role, such as lingual lipase activity and co-lipase deficiency (15,16). The finding of low duodenal bicarbonate output after secretin stimulation indicates a discrepancy between acinar (enzymes) and ductular (fluid, bicarbonate, electrolytes) pancreatic function, not often seen in other forms of pancreatic insufficiency (2). The CF patients, regardless of their ability to produce adequate pancreatic enzyme output, have an abnormality of pancreatic fluid and bicarbonate secretion related to the genetically defective chloride channel function (17, 18). To determine this discrepancy between acinar and ductular function, both secretin and cholecystokinin stimulation is necessary in pancreatic function testing. All CF patients had evidence of an impaired endocrine pancreatic function as reflected by a blunted insulin secretion and virtual absence of PP secretion in response to a standard meal. This finding is in agreement with a previous study in adult CF patients (19). PP-producing cells occur almost exclusively in the pancreas (20). Postprandial increase in plasma PP has been suggested to be a measure of functional pancreatic mass in chronic pancreatitis, but in these patients the PP secretion remained normal until the pancreas was so damaged that the disease usually had revealed itself by overt steatorrhea (20). PP cells are scattered in the exocrine parenchyma and in the islets, where they are usually located at the periphery (20,21). In the present study basal and maximal plasma PP levels correlated well with the diamter of the pancreatic body, suggesting that pancreatic atrophy is accompanied by a decrease in either the number or function of PP cells. Disruption of pancreatic architecture by
Scand J Gastroenterol Downloaded from informahealthcare.com by TIB/UB Hannover on 01/10/15 For personal use only.
698
E. H . J . van Haren et al.
fibrosis and fatty infiltration is probably responsible for the observed findings (13). A blunted but still significant insulin response to meal ingestion was found irrespective of exocrine pancreatic function. The results therefore indicate that the insulin-producing cells located in the islets are much less susceptible to the damaging process of the pancreas in CF than the PP-producing cells and the exocrine tissue. Nevertheless, the number of islets per cubic millimeter of pancreatic tissue was decreased in CF patients with and without diabetes (22). In addition, it should be noted that islet cell antibodies may also play a role in the development of diabetes mellitus in CF (23). The sweat test is the cornerstone in the diagnosis of CF, but the test can yield borderline results. Modification of the sweat test by means of fluorocortisone can aid the diagnosis further (24), but normal sweat electrolyte levels may be found in patients with typical CF (25). In such cases ancillary criteria for the diagnosis of CF have to be relied on. The demonstration of characteristic features at ultrasound and at exocrine pancreatic function testing can support the diagnosis of CF. It is concluded that in adult CF patients ultrasonography demonstrates abnormalities in pancreatic morphology which are characterized by a large pancreatic head relative to a small and echo-dense body. Exocrine and endocrine pancreatic function are disturbed in all patients, irrespective of the presence of steatorrhea. These functional and morphologic abnormalities can support the diagnosis of CF in case of borderline or negative sweat test results.
REFERENCES 1. David TJ. Cystic fibrosis. Arch Dis Child 1990;65:152-7. 2. Park RW, Grand RJ. Gastrointestinal manifestations of cystic fibrosis: a review. Gastroenterology 1981;81:114M1. 3. Hall SK, Stableforth DE, Green A. Sweat sodium and chloride eoneentrations-essential criteria for the diagnosis of cystic fibrosis in adults. Ann Clin Biochem 1990;27:318-20. 4. Niederau C, Sonnenberg A, Muller JE, Eckenbrecht JF, Scholten T, Fritsch WP. Sonographic measurements of the normal liver, spleen, pancreas and portal vein. Radiology 1Y83; 149:53740. 5. Yalow RS, Bcrson SA. Immunoassay of endogenous plasma insulin in man. J Clin Invest 1960;39:1157-75. 6. Lamers CBHW, Diemel CM, Jansen JBMJ. Comparative study of plasma pancreatic polypeptide responses to food, secretin
Received 27 December 1991 Accepted 3 March 1992
and bombesin in normal subject and in patients with chronic pancreatitis. Dig Dis Sci 1984;29: 102-8. 7. Bratton AC, Marshall EK. A new coupling component for sulfonimide determination. J Biol Chem 1939;128:537-50, 8. Lankisch PG. Exocrine pancreatic function tests. Gut 1982; 23: 777-98. 9. MacCarthy RL, Word EM, Charboncau JW, Sheedy PF, May GR. Imaging of the pancreas. In: Go VLW, et al., editors. The exocrine pancreas: biology, pathobiology and diseases. New York: Raven Prcss, 1986:841-72. 10. Graham N , Manhire AR, Stead RJ, Lees WR, Hodson ME, Batten JC. Cystic fibrosis: ultrasonographic findings in the pancreas and hepatobiliary system correlated with clinical data and pathology. Clin Radio1 1985;36: 199-203. 11. Phillips HE, Cox CL, Reid MH, McGahan JP. Pancreatic sonography in cystic fibrosis. AJR 1981:137:6Y-72. 12. Schwachrnan 11, Lebenthal E, Khaw KT. Recurrent acute pancreatitis in patients with cystic fibrosis with normal pancreatic enzymes. Pediatrics 1975;55:86-95. 13. Lebenthal E, Lerner A, Heitlinger L. The pancreas in cystic fibrosis. In: Go VLW, et al., editors. The exocrine pancreas: biology, pathobiology and diseases. New York: Raven Press, 1986:783-X 17. 14. Hernanz M , Teele RL, Perez A, et al. Pancreatic cystosis i n cystic fibrosis. Radiology 1986;158:629-31. IS. Fredrikson B, Blackberg L. Lingual lipase: an important lipase in the digestion of dietary lipids in cystic fibrosis. Pediatr Rcs 1980; 14: 1387-90. 16. Gaskin KJ, Durie PR, Lee L. Colipase and lipase secretion in childhood-onset pancreatic insufficiency. Gastroenterology 1984;86:1-7. 17. Gaskin KJ, Durie PR, Corey M, Wei P, Forstner G. Evidence for a primary defcct of pancreatic HCO,-secretion in cystic fibrosis. Pcdiatr Res 1982;16:554-7. 18. Davidson GP, Kirubakaran CP, Ratcliffe G , Cooper DM, Robb TA. Abnormal pancreatic electrolyte secretion in cystic fibrosis. Acta Pacdiatr Scand 1986;75:145-50. 19. Allen JM, Penketh ARL, Adrian TE, et al. Adult cystic fibrosis: postprandial rcsponse of gut regulatory peptides. Gastroenterology 1983;85:1379-83. 20. Schwartz TW. Pancreatic polypeptide: a hormone under vagal control. Gastroenterology 1983;85: 1411-25. 21. Larsson L1, Sundler F, Hakanson R. Pancreatic polypeptidea postulated new hormone: identification of its cellular storage site by light and electron microscopic immunocytochcrnistry. Diabetologia 1976; 12:211-26. 22. Rodman HA, Docrshuli CF, Roland JM. The interaction of two diseases: diabetes mellitus and cystic fibrosis. Medicine 1986;65:389-97. 23. Stutchlield PR, O’Halloran SM, Smith CS, Woodrow JC, Bottazzo GF, Heaf 0. HLA-typc, islet cell antibodies and glucose intolerance in cystic fibrosis. Arch Dis Child 1988;63: 1234-9. 24. Hodson ME, Beldon J, Power R , Duncan F. Bambcr M, Batten JC. Sweat tests to diagnose cystic fibrosis in adults. Br Med J 1983i286: 1381-3. 25. Davis PB, Hubbard VS, Di Sant’Agnese PA. Low sweat electrolytes in a patient with cystic fibrosis. Am J Med 1980; 69:643-6.
